Determination of Inverse Depths Using Direct Boussinesq Modeling

This paper presents a technique to reconstruct bathymetry using two snapshots of water surface elevations and velocities as data. A Boussinesq model initialized with the data from the first snapshot is used as the engine to compute wave evolution over test bathymetries, which are iterated until a best fit is reached with the second snapshot. Phase speed difference in computed and measured data is used as the basis for updating bathymetry at each iteration. A novel technique is used to minimize the effect of mismatches between velocity and surface elevation, which could otherwise result in significant errors in phase speed estimates. The inversion algorithm is found to reconstruct bathymetries well for a variety of test cases. A particular strength of the methodology is the ability to account consistently for strong, unsteady currents while constructing the inverse bathymetry.